EP0237555B1 - Method and device for cutting the contour of a plane object - Google Patents

Method and device for cutting the contour of a plane object Download PDF

Info

Publication number
EP0237555B1
EP0237555B1 EP19860905837 EP86905837A EP0237555B1 EP 0237555 B1 EP0237555 B1 EP 0237555B1 EP 19860905837 EP19860905837 EP 19860905837 EP 86905837 A EP86905837 A EP 86905837A EP 0237555 B1 EP0237555 B1 EP 0237555B1
Authority
EP
European Patent Office
Prior art keywords
image
contour
cutting
data
point
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired
Application number
EP19860905837
Other languages
German (de)
French (fr)
Other versions
EP0237555A1 (en
Inventor
Charlie Krey
Alain Ayache
André Bruel
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Centre National de la Recherche Scientifique CNRS
Original Assignee
Centre National de la Recherche Scientifique CNRS
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Family has litigation
First worldwide family litigation filed litigation Critical https://patents.darts-ip.com/?family=9323678&utm_source=google_patent&utm_medium=platform_link&utm_campaign=public_patent_search&patent=EP0237555(B1) "Global patent litigation dataset” by Darts-ip is licensed under a Creative Commons Attribution 4.0 International License.
Application filed by Centre National de la Recherche Scientifique CNRS filed Critical Centre National de la Recherche Scientifique CNRS
Publication of EP0237555A1 publication Critical patent/EP0237555A1/en
Application granted granted Critical
Publication of EP0237555B1 publication Critical patent/EP0237555B1/en
Expired legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B19/00—Programme-control systems
    • G05B19/02—Programme-control systems electric
    • G05B19/42—Recording and playback systems, i.e. in which the programme is recorded from a cycle of operations, e.g. the cycle of operations being manually controlled, after which this record is played back on the same machine
    • C—CHEMISTRY; METALLURGY
    • C14—SKINS; HIDES; PELTS; LEATHER
    • C14B—MECHANICAL TREATMENT OR PROCESSING OF SKINS, HIDES OR LEATHER IN GENERAL; PELT-SHEARING MACHINES; INTESTINE-SPLITTING MACHINES
    • C14B5/00—Clicking, perforating, or cutting leather

Definitions

  • the invention made in the laboratory designated "Languages and Computer Systems" of the National Polytechnic Institute of Toulouse, unit associated with the C.N.R.S. number 347, relates to a method for determining the course of the contour cutting of a plane object as a function of parameters relating to the peripheral appearance of said object. This process aims, automatically, to analyze the shape of a cutting object, to deduce from this analysis the coordinates of the cutting plot and, finally, to pilot an industrial machine carrying out the actual cutting.
  • This invention extends a device for implementing said method.
  • the invention can in particular be applied to carry out the operation known as "contouring" of the skins in the field of tanning.
  • This contouring operation consists in making the skins which, after having undergone the essential mechanical and chemical treatments, have more irregularities around their edges, fraying, tears, fringes ... These irregularities are in fact unusable from an industrial point of view, although billed upon purchase.
  • this task is carried out by means of various industrial devices (scissors, electric scissors, fluid jet, laser ...) controlled manually. Cutting is therefore a function of the worker's technique and also of his fatigue and constitutes long and laborious work.
  • the present invention proposes to overcome the shortcomings of existing manual cutting systems.
  • the essential objective of the invention is to provide a system capable of automatically analyzing the shape of a plane object and of deducing therefrom the optimal path for cutting, according to standardized rules.
  • Another objective of the invention is to provide a system benefiting from a great flexibility of adaptation in order to be able to vary the cutting parameters at the discretion of the user (dimensions of the irregularities to be eliminated, distance from the periphery of the holes to contour, distance between rough contour and cut).
  • Another objective is to provide a simple structure system capable of allowing high processing rates.
  • the process targeted by the invention therefore comprises two successive stages, clearly differentiated.
  • the first step intended to smooth the contour, consists in digitally processing the video signal representative of a portion of the contour of the object, so as to determine the coordinates of the plot of the cutout for each image.
  • the second stage memorizes the coordinates of the route and makes it possible to take care of the problems of continuity of the route and the choice of the movement of the camera.
  • the process targeted by the invention makes it possible to complete the cutting path corresponding to a given image with a portion of the outline of the following image and to move the camera to digitize this new image with an optimal position.
  • Image binarization allows the use of calculation methods applying to binary images. This binarization is carried out by comparing the gray level of each point of the image with a threshold whose gray level is fixed either experimentally or during the initialization of the system by taking the image of a known object. This method thus makes it possible to separate the points of the background from the points belonging to the object, the outline of the latter then being defined by a transition between a point belonging to said object and a point belonging to the background.
  • the tests carried out on said information consisting in comparing the position of the start point on and respectively of the end point Fn of said image (n) with stored reference values whose position is determined, with a view to delivering information representative of the position from the end point Fn of the image (n), in the image (n + 1).
  • the method consists in ordering said coordinates according to the direction of travel chosen and in generating an automatic movement of the camera which ensures continuity in the analysis of the skin contour and in determining the cutting path from one image to the next.
  • This movement of the camera could have been systematic by placing the end point Fn of the outline of the image (n) in the center of the next image (n + 1).
  • this positioning which involves a new analysis of half of the previous image, would noticeably slow down the process.
  • the process targeted by the invention therefore proposes to find an optimal positioning of the end point Fn of the contour of the image (n), in the following image (n + 1), and therefore an optimal positioning of the camera.
  • the method consists, according to another characteristic of the invention, of defining a determined number of reference points whose position is predetermined and of partitioning each image into a determined number of regions, the tests carried out on the information representative of the position of the start point on and the end point Fn of the contour portion of an image (n) consisting, on the one hand, in determining the reference point whose distance to the start point on is minimal, d ' on the other hand, to determine the region of the image (n) where the end point Fn is located, and to deduce therefrom, according to programmed criteria, the position of said end point Fn in the contiguous image (n + 1) .
  • end point Fn of the image (n) is generally positioned on one of the reference points so as to constitute the start point (Dn + 1) of the contour portion of the image (n + 1). .
  • the method makes it possible to define the position of the point Fn of this image (n) in the image ( n + 1), so that the portion of the contour already processed in the image (n) and also belonging to the image (n + 1) is minimal, and the portion of the contour processed in the image (n + 1) is maximum.
  • a vision system mainly consisting of a video camera, an analog / digital conversion unit, an analysis processor suitable for processing digital codes, and d a decision processor adapted to process the information transmitted by the analysis processor is known per se.
  • Such systems make it possible to generate a video image of an object and to process by calculation the analog signal of this image according to determined criteria.
  • An exemplary embodiment is described in particular in French patent No. 2,543,457 which describes a vision system allowing the sorting of fruits according to their external appearance.
  • the present invention is characterized not simply by the implementation of a vision system as described above but by the combination of this system with means for moving it and means for controlling said means of displacements adapted to ensure continuity in the analysis of the outline of a plane object and in determining the path of the cut, from one image to the next.
  • the device shown by way of example in FIG. 1 comprises a table 1 with planar displacement X, Y provided with a bracket 2 able to move longitudinally relative to said table.
  • This bracket 2 is further provided with a support 3, movable relative to the latter, in a direction perpendicular to its direction of movement.
  • This table also includes drive means consisting of two stepper motors 4 adapted to drive the stem 2 and the support 3 independently of each other along orthogonal directions.
  • a video camera 5 On the support 3 is arranged a video camera 5, the height of which above the displacement table 1 is adjustable so as to be able to modify the field of vision of this camera as a function of the level of precision required.
  • This camera is preferably of the "C.C.D” charge transfer type, so as to allow real-time capture of the contour of the object during continuous movement of the camera.
  • This plan object 6 placed on the displacement table can be made up of a skin whose outline it is desired to harmonize (elimination of fraying, tears, fringes) after having subjected it to conventional mechanical and chemical treatments.
  • This object 6 can also consist of a plate on which is represented a logo or any other motif which it is desired to cut.
  • the video camera 5 delivers, in a conventional manner, an analog video signal, representative of a contour portion of this object 6 to an analog / digital conversion unit 7.
  • This video signal is captured, converted in the form of a matrix of numerical codes depending on the luminance level of said sampled signal, then processed in an analysis processor 8 to deliver a series of information representative of the plot of the cut.
  • This series of information representative of the plot of the cut from the analysis processor 8 is delivered through transmission means 9 to a pilot processor 10.
  • This processor 10 is adapted to memorize the position of said points of the plot and to perform tests established according to programmed criteria with a view to delivering information allowing the control of the camera for viewing the portion of contiguous contour.
  • the pilot processor 10 delivers the control information from the camera 4 to the drive means 11 of the supports 2 - 3 of this camera and also delivers a synchronization signal to the converter 7 authorizing the entry of a new image.
  • the analog / digital conversion unit 7 receives the video signal from the camera 5 and the synchronization signal from the pilot processor 10. This unit 7 therefore captures the video signal from the image contiguous to the previously processed image and affects at the points of this image of the multi-bit codes representative of the gray levels of these points among a number of predetermined levels between white and black, thus delivering a matrix of digital codes representative of said image.
  • This matrix of digital codes is delivered to the analysis processor 8 comprising a storage memory 12 capable of storing said matrix of digital codes on command of a direct memory access unit 13, a calculation unit 14 composed of a micropro interrupter, program memories 15 and a working memory 16 associated with the microprocessor.
  • the program memories 15 contain the program of the successive operations to be carried out by the microprocessor from the matrix of digital codes contained in the storage memory 12.
  • the first operation consists in comparing the gray level of each numerical code with the gray level of a determined threshold, either experimentally or during the initialization of the system by generating the image of a known object. This operation leads to obtaining a binary coded matrix from which it is easy to determine the points of the image belonging to the background and the points of the image belonging to the object.
  • the second operation consists in smoothing the outline of the object.
  • This smoothing operation consists in treating all the points of the image by morphological methods such as: erosion and dilation, making it possible to harmonize the contour by removing the irregularities from it.
  • erosion and expansion methods are easily adaptable to the cutting requirements of each user without modifying the structure of the device. Indeed, the characteristics of the outline of the cut essentially depend on the number of erosions and expansions to which the points belonging to the outline of the object are subjected.
  • the following operation consists in extracting the position of the points representative of the plot of the cut and in ordering the ordering of said points in order to restore them in the order of travel of the contour.
  • the extraction of the points of the contour consists in determining the points of which at least one neighboring point on the same line or the same column belongs to the background.
  • Their scheduling consists in carrying out the follow-up of the plot of the cutout starting from the edges of the image and in following said plot up to another limit of the image. This operation is repeated as long as there are untreated points of the plot, at the edge of the image.
  • the unmarked contours represent holes located inside the object, at a distance from the edge of this skin large enough that the successive erosion and expansion operations have not eliminated it.
  • each matrix of digital codes is carried out in real time during the period of time separating the end of their storage in the storage memory 12 and the capture of a new image in the conversion unit 7
  • the processing time of an image performing binarization, then three erosions and dilations, then the contour extraction and finally the tracking of this contour is of the order of three seconds.
  • the various binarization, erosion, expansion, contour extraction operations can be carried out by means of wired modules, each used specifically for a specific operation, which allow the calculation times to be further reduced.
  • wired modules are arranged in series and carry out the processing of the digital codes delivered by the analog / digital conversion unit 7, so as to deliver to the pilot processor 10 information representative of the plot of the cut.
  • a device fitted with wired modules therefore does not have an analysis processor, and the information on the cutting plot is transmitted directly to the pilot processor which in this case has a storage memory capable of storing said information on command of a direct memory access unit.
  • the information representative of the cutting plot resulting from the contour tracking after image processing is then delivered to the pilot processor 10 whose role is to ensure continuity in determining the cutting path, from an image to the other, by completing the image cutting path (n) with a portion of the image outline (n + 1
  • This pilot processor 10 receives a series of information which is distributed via a data bus 17 to a computing unit 18 consisting of a microprocessor; this is associated with a program memory 19 containing the tests to be carried out and a working memory 20.
  • the present invention proposes to define five benchmarks whose position is determined ( Figure 3a).
  • Four of these five points PN, PE, PS and PW are located on the periphery of the image, each in the center of one of the sides of said image.
  • the fifth point PC is positioned in the center of the image.
  • the role of the pilot processor 10 consists in controlling a movement of the camera 5 so that the end point Fn of the image (n) is positioned on one of the reference points PN, PE, PS, PW or PS in the following image (n + 1), so that the portion of the contour already processed in the image (n) and also belonging to the image (n + 1) is minimal and the portion of the contour processed in the image (n + 1) maximum.
  • the start point D of the contour of each image is positioned on a reference point PN, PE, PS, PW, PC.
  • the pilot processor 10 determines which of said contours is connected to the image (n) by searching for the one which has a point positioned on or near the point of coordinate system on which the end point Fn of the image (n) is located in the image (n + 1).
  • Each image is also partitioned into nine similar regions R1 ... R9 (FIG. 3b) which make it possible to take into account the position of the end point Fn of the contour in the image (n).
  • the control of the camera with a view to viewing an image (n + 1) therefore takes account of the position of the initial point on and respectively of the end point Fn of the cutting path in the image (n) and consists in determining , from said positions and according to memorized criteria, the reference point on which the end point Fn will be positioned in the image (n +1
  • This method makes it possible, with judiciously defined selection criteria, to reduce by around 30% the number of images necessary for capturing the entire outline of an object compared to a strategy systematically placing the end point Fn at the center of the 'image (n + 1).
  • the pilot processor stores the coordinates in X and Y of the camera 5 on the movement table 1 and as a function of the coordinates of the starting point and the direction of travel, can thus determine, when viewing an "invalid" image ", whether or not this image corresponds to a particular region of the skin.
  • the processor inhibits the results of the tests taking into account the positions of the start point and the end point of the outline of the image (n) and determines the position of the end point of the image (n) in l image (n + 1) depending on the region where the camera is positioned.
  • FIG. 4 represents for this purpose all the combinations making it possible to determine the position of the end point Fn of the image (n), in the image (n + 1).
  • the information stored in the working memory 20 of the pilot processor 10 is delivered via communication interfaces 21 to industrial cutting means 22 such as electric scissors, laser , jet of fluid, capable of carrying out the actual cutting of the skin.
  • FIGS. 5a and 5b represent a skin before and after treatment and it can be noted that this skin has, after treatment, a much more regular outline.
  • a device as described above, and using wired modules in order to carry out the binarization, erosion, expansion, contour extraction operations, makes it possible to capture the outline of a sheep skin, and to determine the optimal cutting path with a rate of two to three skins per minute. This performance can also be improved by having two mobile cameras, each analyzing part of the skin.

Landscapes

  • Engineering & Computer Science (AREA)
  • Mechanical Engineering (AREA)
  • Chemical & Material Sciences (AREA)
  • Organic Chemistry (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Image Analysis (AREA)
  • Image Processing (AREA)

Abstract

Method and device for determining the cutting path of the contour of a plane object as a function of the parameters relating to the peripheral aspect of said object. Said method comprises the generation of a video image of a portion of the object contour by means of a video camera, the conversion of the analog signal of said image into a matrix of digital codes representative of the grey level of the dots, the computer processing of said matrix as a function of the cutting parameters so as to deliver information representative of the cutting layout, the storing of said information, performing a series of tests set up according to programmed criteria on the information in order to provide information enabling piloting of the camera for viewing a contiguous contour portion and controlling the effective cutting when all the contour has been stored.

Description

L'invention, faite dans le Laboratoire désigné "Langages et Systèmes Informatiques" de l'Institut National Polytechnique de Toulouse, unité associée au C.N.R.S. numéro 347, concerne un procédé de détermination du tracé de la découpe du contour d'un objet-plan en fonction de paramètres concernant l'aspect périphérique dudit objet. Ce procédé vise, de façon automatique, analyser la forme d'un objet découper, déduire de cette analyse les coordonnées du tracé de la découpe et, enfin piloter une machine industrielle réalisant la découpe effective. Cette invention s'étend un dispositif pour la mise en oeuvre dudit procédé.The invention, made in the laboratory designated "Languages and Computer Systems" of the National Polytechnic Institute of Toulouse, unit associated with the C.N.R.S. number 347, relates to a method for determining the course of the contour cutting of a plane object as a function of parameters relating to the peripheral appearance of said object. This process aims, automatically, to analyze the shape of a cutting object, to deduce from this analysis the coordinates of the cutting plot and, finally, to pilot an industrial machine carrying out the actual cutting. This invention extends a device for implementing said method.

L'invention peut être en particulier appliquée pour réaliser l'opération dite de "contournage" des peaux dans le domaine de la mégisserie. Cette opération de contournage consiste rendre plus esthétique les peaux qui, après avoir subi les traitements mécaniques et chimiques indispensables, présentent des irrégularités sur leur pourtour effilochures, déchirures, franges... Ces irrégularités sont en effet inexploitables du point de vue industriel, bien que facturées lors de l'achat. A l'heure actuelle, cette tâche est réalisée au moyen d'appareils industriels divers (ciseaux, ciseaux électriques, jet de fluide, laser...) commandés manuellement. La découpe est donc fonction de la technique de l'ouvrier et également de sa fatigue et constitue un travail long et laborieux.The invention can in particular be applied to carry out the operation known as "contouring" of the skins in the field of tanning. This contouring operation consists in making the skins which, after having undergone the essential mechanical and chemical treatments, have more irregularities around their edges, fraying, tears, fringes ... These irregularities are in fact unusable from an industrial point of view, although billed upon purchase. At present, this task is carried out by means of various industrial devices (scissors, electric scissors, fluid jet, laser ...) controlled manually. Cutting is therefore a function of the worker's technique and also of his fatigue and constitutes long and laborious work.

La présente invention se propose de pallier les défauts des systèmes de découpe manuelle existants.The present invention proposes to overcome the shortcomings of existing manual cutting systems.

L'objectif essentiel de l'invention est de fournir un système capable d'analyser automatiquement la forme d'un objet-plan et d'en déduire le chemin optimal pour la découpe, en fonction de règles normalisées.The essential objective of the invention is to provide a system capable of automatically analyzing the shape of a plane object and of deducing therefrom the optimal path for cutting, according to standardized rules.

Un autre objectif de l'invention est de fournir un système bénéficiant d'une grande souplesse d'adaptation en vue de pouvoir faire varier les paramètres de la découpe au gré de l'utilisateur (dimensions des irrégularités éliminer, distance la périphérie des trous à contourner, distance entre contour brut et découpe).Another objective of the invention is to provide a system benefiting from a great flexibility of adaptation in order to be able to vary the cutting parameters at the discretion of the user (dimensions of the irregularities to be eliminated, distance from the periphery of the holes to contour, distance between rough contour and cut).

Un autre objectif est de fournir un système de structure simple, apte permettre des cadences de traitement élevées.Another objective is to provide a simple structure system capable of allowing high processing rates.

A cet effet, le procédé visé par l'invention, de détermination du tracé de la découpe du contour d'un objet-plan en-fonction de paramètres concernant l'aspect périphérique dudit objet est du type consistant:

  • - à générer, au moyen d'une caméra vidéo une image tramée d'une portion du contour de l'objet constituée de points répartis en lignes et en colonnes,
  • - à saisir en temps réel le signal analogique de cette image, le convertir en une matrice de codes numériques, chacun représentatif du niveau de gris d'un point de l'image et
  • mémoriser cette matrice en temps réel. Selon la présente invention, ce procédé consiste:
  • - à traiter, par le calcul, la matrice de codes numériques en fonction des paramètres de la découpe, de façon à déterminer les points représentatifs du tracé de la découpe,
  • - à mémoriser des informations représentatives de la position des points représentatifs du tracé de la découpe,
  • - à effectuer sur lesdites informations une série de tests établis selon des critères programmés en vue de délivrer des informations permettant le pilotage de la caméra pour la visualisation de la portion de contour contiguë,
  • - et à commander la découpe effective lorsque la totalité du contour est mémorisée.
To this end, the process targeted by the invention, for determining the layout of the cutting of the outline of a plane object as a function of parameters relating to the peripheral appearance of said object is of the type consisting:
  • to generate, by means of a video camera, a raster image of a portion of the outline of the object consisting of points distributed in rows and columns,
  • - to capture the analog signal of this image in real time, convert it into a matrix of digital codes, each representative of the gray level of a point in the image and
  • memorize this matrix in real time. According to the present invention, this method consists:
  • to process, by calculation, the matrix of digital codes as a function of the cutting parameters, so as to determine the points representative of the cutting plot,
  • - memorizing information representative of the position of the points representative of the plot of the blank,
  • to carry out on said information a series of tests established according to programmed criteria with a view to delivering information allowing the control of the camera for viewing the portion of contiguous contour,
  • - and to control the actual cutting when the entire contour is memorized.

Le procédé visé par l'invention comporte donc deux étapes successives, bien différenciées. La première étape, destinée réaliser le lissage du contour, consiste réaliser le traitement numérique du signal vidéo représentatif d'une portion du contour de l'objet, de façon déterminer les coordonnées du tracé de la découpe pour chaque image. La deuxième étape réalise la mémorisation des coordonnées du tracé et permet de prendre en charge les problèmes de continuité du tracé et le choix du déplacement de la caméra. Dans ce but, le procédé visé par l'invention permet de compléter le chemin de la découpe correspondant à une image donnée avec une portion de contour de l'image suivante et de déplacer la caméra pour numériser avec une position optimale cette nouvelle image.The process targeted by the invention therefore comprises two successive stages, clearly differentiated. The first step, intended to smooth the contour, consists in digitally processing the video signal representative of a portion of the contour of the object, so as to determine the coordinates of the plot of the cutout for each image. The second stage memorizes the coordinates of the route and makes it possible to take care of the problems of continuity of the route and the choice of the movement of the camera. To this end, the process targeted by the invention makes it possible to complete the cutting path corresponding to a given image with a portion of the outline of the following image and to move the camera to digitize this new image with an optimal position.

Selon une caractéristique de l'invention, la première étape de détermination du tracé de la découpe consiste:

  • - à numériser la matrice de codes numériques sous forme de codes binaires dont la valeur est fonction de la présence ou de l'absence d'objet en vue d'obtenir des informations représentatives du contour de l'objet,
  • - à traiter lesdites informations représentatives du contour de l'objet par des méthodes morphologiques telles que érosion, dilatation... en vue de délivrer des informations représentatives de la position des points du tracé de la découpe.
According to a characteristic of the invention, the first step of determining the layout of the cut consists:
  • - to digitize the matrix of digital codes in the form of binary codes whose value depends on the presence or absence of an object in order to obtain information representative of the outline of the object,
  • - to process said information representative of the outline of the object by morphological methods such as erosion, expansion, with a view to delivering information representative of the position of the points of the plot of the cut.

La binarisation de l'image permet l'utilisation de méthodes de calcul s'appliquant à des images binaires. Cette binarisation est effectuée en comparant le niveau de gris de chaque point de l'image avec un seuil dont le niveau de gris est fixé soit expérimentalement, soit lors de l'initialisation du système en prenant l'image d'un obiqt connu. Cette méthode permet ainsi de séparer les points du fond des points appartenant l'objet, le contour de celui-ci étant alors défini par une transition entre un point appartenant audit objet et un point appartenant au fond.Image binarization allows the use of calculation methods applying to binary images. This binarization is carried out by comparing the gray level of each point of the image with a threshold whose gray level is fixed either experimentally or during the initialization of the system by taking the image of a known object. This method thus makes it possible to separate the points of the background from the points belonging to the object, the outline of the latter then being defined by a transition between a point belonging to said object and a point belonging to the background.

Le traitement de cette matrice de codes binaires au moyen de méthodes morphologiques permet d'obtenir un lissage du contour de l'objet dont les spécifications fonctions du nombre d'opérations d'érosions ou de dilatations peuvent être modifiées au gré de l'utilisateur.Processing this matrix of binary codes using morphological methods makes it possible to obtain a smoothing of the outline of the object. whose specifications depending on the number of erosion or expansion operations can be modified at the user's discretion.

Cette première étape du traitement de l'image vidéo ayant été réalisée, le procédé consiste alors pour chaque image (n):

  • - à procéder l'ordonnancement des informations mémorisées, représentatives de la position des points du tracé de la découpe en vue de les restituer dans l'ordre de parcours du contour,
  • - à extraire desdites informations les informations représentatives de la position du point de début on de la portion de contour, et du point final fn de ladite portion de contour,
This first step of processing the video image having been carried out, the method then consists for each image (n):
  • - proceed with the ordering of the stored information, representative of the position of the points of the plot of the cutout in order to restore them in the order of travel of the contour,
  • extracting from said information the information representative of the position of the start point on of the contour portion, and of the end point fn of said contour portion,

les tests effectués sur lesdites informations consistant à comparer la position du point de début on et respectivement du point final Fn de ladite image (n) avec des valeurs de référence mémorisées dont la position est déterminée, en vue de délivrer une information représentative de la position du point final Fn de l'image (n), dans l'image (n+1).the tests carried out on said information consisting in comparing the position of the start point on and respectively of the end point Fn of said image (n) with stored reference values whose position is determined, with a view to delivering information representative of the position from the end point Fn of the image (n), in the image (n + 1).

Ainsi les coordonnées du tracé de la découpe d'une image (n) ayant été déterminées, le procédé consiste à ordonner lesdites coordonnées suivant le sens de parcours choisi et à engendrer un déplacement automatique de la caméra qui assure la continuité dans l'analyse du contour de la peau et dans la détermination du chemin de découpe d'une image à la suivante.Thus the coordinates of the outline of the cutting of an image (n) having been determined, the method consists in ordering said coordinates according to the direction of travel chosen and in generating an automatic movement of the camera which ensures continuity in the analysis of the skin contour and in determining the cutting path from one image to the next.

Ce déplacement de la caméra aurait pu être systématique en plaçant le point final Fn du contour de l'image (n) au centre de l'image suivante (n+1). Toutefois, ce positionnement qui entraîne une nouvelle analyse de la moitié de l'image précédente ralentirait notablement le processus. Le procédé visé par l'invention se propose donc de trouver un positionnement optimal du point final Fn du contour de l'image (n), dans l'image suivante (n+1), et donc un positionnement optimal de la caméra.This movement of the camera could have been systematic by placing the end point Fn of the outline of the image (n) in the center of the next image (n + 1). However, this positioning, which involves a new analysis of half of the previous image, would noticeably slow down the process. The process targeted by the invention therefore proposes to find an optimal positioning of the end point Fn of the contour of the image (n), in the following image (n + 1), and therefore an optimal positioning of the camera.

A cet effet, le procédé consiste, selon une autre caractéristique de l'invention, définir un nombre déterminé de points de repère dont la position est prédéterminée et à partitionner chaque image en un nombre de régions déterminées, les tests effectués sur les informations représentatives de la position du point de début on et du point final Fn de la portion de contour d'une image (n) consistant, d'une part, à déterminer le point de repère dont la distance au point de début on est minimale, d'autre part, à déterminer la région de l'image (n) où se trouve positionné le point final Fn, et à en déduire, en fonction de critères programmés, la position dudit point final Fn dans l'image contiguë (n + 1).To this end, the method consists, according to another characteristic of the invention, of defining a determined number of reference points whose position is predetermined and of partitioning each image into a determined number of regions, the tests carried out on the information representative of the position of the start point on and the end point Fn of the contour portion of an image (n) consisting, on the one hand, in determining the reference point whose distance to the start point on is minimal, d ' on the other hand, to determine the region of the image (n) where the end point Fn is located, and to deduce therefrom, according to programmed criteria, the position of said end point Fn in the contiguous image (n + 1) .

En outre, le point final Fn de l'image (n) est généralement positionné sur un des points de repère de façon à constituer le point de début (Dn+1) de la portion de contour de l'image (n + 1).In addition, the end point Fn of the image (n) is generally positioned on one of the reference points so as to constitute the start point (Dn + 1) of the contour portion of the image (n + 1). .

Ainsi, les positions respectives du point de début et du point final du tracé de la découpe d'une image (n) ayant été déterminées, le procédé permet de définir la position du point Fn de cette image (n) dans l'image (n + 1), de sorte que la portion de contour déjà traitée dans l'image (n) et appartenant aussi à l'image (n+1) soit minimale, et que la portion de contour traitée dans l'image (n + 1) soit maximale.Thus, the respective positions of the start point and the end point of the plot of the cutting of an image (n) having been determined, the method makes it possible to define the position of the point Fn of this image (n) in the image ( n + 1), so that the portion of the contour already processed in the image (n) and also belonging to the image (n + 1) is minimal, and the portion of the contour processed in the image (n + 1) is maximum.

Un choix judicieux du point de repère pour le positionnement dans l'image (n + 1) du point final Fn du contour de l'image (n) permet ainsi de diminuer d'environ 30 % le nombre de vues nécessaires à la visualisation de la totalité du contour, par rapport à un procédé consistant à placer systématiquement ce point Fn au centre de l'image suivante.A judicious choice of the reference point for positioning in the image (n + 1) of the end point Fn of the contour of the image (n) thus makes it possible to reduce by around 30% the number of views necessary for the visualization of the entire contour, with respect to a process consisting in systematically placing this point Fn in the center of the following image.

L'invention s'étend à un dispositif pour la découpe du contour d'un objet-plan en vue de la mise en oeuvre du procédé ci-dessus défini. Ce dispositif est du type comprenant:

  • - au moins une caméra vidéo agencée pour générer successivement des images vidéo de portions du contour de l'objet,
  • - une unité de conversion analogique/numérique recevant le signal vidéo de la caméra et adaptée pour assurer la saisie de ce signal et délivrer une matrice de codes numériques fonctions du niveau de luminance du signal vidéo échantillonné,
  • - un module dit "processeur d'analyse" comprenant:
    • . une mémoire de stockage apte à mémoriser chaque matrice de codes numériques et commandée par une unité d'accès direct mémoire apte à gérer le rangement en mémoire de codes numériques,
    • . une unité de calcul adaptée pour assurer le traitement des codes numériques contenus dans la mémoire de stockage et associée d'une part à une mémoire programme contenant un programme d'opérations logiques à effectuer, d'autre part, à une mémoire de travail pour le stockage provisoire des résultats,
  • - un module dit "processeur pilote" agencé pour recevoir les informations issues du processeur d'analyse et comprenant une unité de calcul adaptée pour effectuer des tests sur les informations reçues et associé, d'une part, à une mémoire-programme contenant les tests à effectuer, d'autre part, à une mémoire de travail pour le stockage des informations issues du processeur d'analyse,
  • - des moyens de transmission aptes à transmettre les informations du processeur d'analyse vers le processeur pilote.
The invention extends to a device for cutting the outline of a plane object with a view to implementing the method defined above. This device is of the type comprising:
  • - at least one video camera arranged to successively generate video images of portions of the outline of the object,
  • - an analog / digital conversion unit receiving the video signal from the camera and adapted to capture this signal and deliver a matrix of digital codes depending on the luminance level of the sampled video signal,
  • - a module called "analysis processor" comprising:
    • . a storage memory capable of memorizing each matrix of digital codes and controlled by a direct memory access unit capable of managing the storage in memory of digital codes,
    • . a calculation unit suitable for processing the digital codes contained in the storage memory and associated on the one hand with a program memory containing a program of logical operations to be carried out, on the other hand, with a working memory for the temporary storage of results,
  • - a module called "pilot processor" arranged to receive information from the analysis processor and comprising a calculation unit adapted to perform tests on the information received and associated, on the one hand, with a program memory containing the tests to be carried out, on the other hand, to a working memory for the storage of information coming from the analysis processor,
  • - Transmission means capable of transmitting the information from the analysis processor to the pilot processor.

Selon la présente invention, ce dispositif comprend en outre:

  • - des moyens de déplacement de chaque caméra aptes à engendrer un déplacement plan desdites caméras,
  • - une interface de communication d'informations, fonctions du résultat des tests, issues du processeur pilote vers les moyens de déplacement de chaque caméra en vue de piloter le déplacement de celles-ci,
  • - une interface de communication des informations concernant le tracé de la découpe issues du processeur pilote vers des moyens de découpe du contour de l'objet-plan.
According to the present invention, this device further comprises:
  • means for moving each camera capable of generating a planar movement of said cameras,
  • - an information communication interface, functions of the test result, from the pilot processor to the means of movement of each camera in order to control the displacement of these,
  • an interface for communicating information concerning the layout of the cut from the pilot processor towards means for cutting the outline of the plan object.

Il est à noter que l'utilisation d'un système de vision constitué principalement d'une caméra vidéo, d'une unité de conversion analogique/ numérique, d'un processeur d'analyse adapté pour assurer le traitement des codes numériques, et d'un processeur de décision adapté pour assurer le traitement des informations transmises par le processeur d'analyse, est connue en soi. De tels systèmes permettent de générer une image vidéo d'un objet et de traiter par le calcul le signal analogique de cette image en fonction de critères déterminés. Un exemple de réalisation est notamment décrit dans le brevet français n° 2 543 457 qui décrit un système de vision permettant le tri de fruits en fonction de leur aspect externe.It should be noted that the use of a vision system mainly consisting of a video camera, an analog / digital conversion unit, an analysis processor suitable for processing digital codes, and d a decision processor adapted to process the information transmitted by the analysis processor is known per se. Such systems make it possible to generate a video image of an object and to process by calculation the analog signal of this image according to determined criteria. An exemplary embodiment is described in particular in French patent No. 2,543,457 which describes a vision system allowing the sorting of fruits according to their external appearance.

Toutefois, la présente invention se caractérise non simplement par la mise en oeuvre d'un système de vision tel que décrit ci-dessus mais par la combinaison de ce système avec des moyens de déplacement de celui-ci et des moyens de pilotage desdits moyens de déplacement adaptés pour assurer la continuité dans l'analyse du contour d'un objet-plan et dans la détermination du chemin de la découpe, d'une image à la suivante.However, the present invention is characterized not simply by the implementation of a vision system as described above but by the combination of this system with means for moving it and means for controlling said means of displacements adapted to ensure continuity in the analysis of the outline of a plane object and in determining the path of the cut, from one image to the next.

D'autres caractéristiques, buts et avantages de l'invention se dégageront de la description qui suit en référence aux dessins annexés, lesquels en présentent un exemple non limitatif sur ces dessins:

  • - la figure 1 est une vue schématique d'un dispositif de détermination du tracé de la découpe du contour d'un objet-plan, conforme à l'invention,
  • - la figure 2 est une vue synoptique partielle de ce dispositif,
  • - les figures 3a, 3b, 3c illustrent les différents paramètres de référence utilisés pour piloter le déplacement de la caméra d'une image à la suivante,
  • - la figure 4 illustre les différentes possibilités de déplacement de la caméra en fonction des paramètres de référence,
  • - les figures 5a et 5b représentent une peau brute avant traitement et un exemple de découpe obtenue.
Other characteristics, objects and advantages of the invention will emerge from the description which follows with reference to the appended drawings, which present a non-limiting example in these drawings:
  • FIG. 1 is a schematic view of a device for determining the course of the cutting of the outline of a plane object, according to the invention,
  • FIG. 2 is a partial block diagram of this device,
  • FIGS. 3a, 3b, 3c illustrate the different reference parameters used to control the movement of the camera from one image to the next,
  • FIG. 4 illustrates the different possibilities for moving the camera as a function of the reference parameters,
  • - Figures 5a and 5b show a raw skin before treatment and an example of cut obtained.

Le dispositif représenté à titre d'exemple à la figure 1 comprend une table 1 à déplacement plan X, Y dotée d'une potence 2 apte à se déplacer longitudinalement par rapport à ladite table. Cette potence 2 est en outre dotée d'un support 3, mobile relativement à celle-ci, selon une direction perpendiculaire à sa direction de déplacement.The device shown by way of example in FIG. 1 comprises a table 1 with planar displacement X, Y provided with a bracket 2 able to move longitudinally relative to said table. This bracket 2 is further provided with a support 3, movable relative to the latter, in a direction perpendicular to its direction of movement.

Cette table comprend également des moyens d'entraînement constitués de deux moteurs 4 pas à pas adaptés pour entraîner indépendamment l'un de l'autre la potence 2 et respectivement le support 3 le long de directions orthogonales.This table also includes drive means consisting of two stepper motors 4 adapted to drive the stem 2 and the support 3 independently of each other along orthogonal directions.

Sur le support 3 est agencée une caméra vidéo 5 dont la hauteur au-dessus de la table de déplacement 1 est réglable de façon à pouvoir modifier le champ de vision de cette caméra en fonction du niveau de précision requis. Cette caméra est de préférence du type à transfert de charges "C.C.D", de façon à permettre une saisie en temps réel du contour de l'objet lors d'un déplacement en continu de la caméra.On the support 3 is arranged a video camera 5, the height of which above the displacement table 1 is adjustable so as to be able to modify the field of vision of this camera as a function of the level of precision required. This camera is preferably of the "C.C.D" charge transfer type, so as to allow real-time capture of the contour of the object during continuous movement of the camera.

Cet objet-plan 6 disposé sur la table de déplacement peut être constitué d'une peau dont on désire harmoniser le contour (suppression des effilochures, déchirures, franges) après lui avoir fait subir les traitements mécaniques et chimiques classiques. Cet objet 6 peut également consister en une plaque sur laquelle est représenté un logo ou tout autre motif dont on désire réaliser la découpe.This plan object 6 placed on the displacement table can be made up of a skin whose outline it is desired to harmonize (elimination of fraying, tears, fringes) after having subjected it to conventional mechanical and chemical treatments. This object 6 can also consist of a plate on which is represented a logo or any other motif which it is desired to cut.

La caméra vidéo 5 délivre, de façon classique, un signal analogique vidéo, représentatif d'une portion de contour de cet objet 6 à une unité de conversion analogique/numérique 7. Ce signal vidéo est saisi, converti sous forme d'une matrice de codes numériques fonctions du niveau de luminance dudit signal échantillonné, puis traité dans un processeur d'analyse 8 pour délivrer une série d'informations représentatives du tracé de la découpe.The video camera 5 delivers, in a conventional manner, an analog video signal, representative of a contour portion of this object 6 to an analog / digital conversion unit 7. This video signal is captured, converted in the form of a matrix of numerical codes depending on the luminance level of said sampled signal, then processed in an analysis processor 8 to deliver a series of information representative of the plot of the cut.

Cette série d'informations représentative du tracé de la découpe issue du processeur d'analyse 8 est délivrée à travers des moyens de transmission 9 vers un processeur pilote 10. Ce processeur 10 est adapté pour mémoriser la position desdits points du tracé et pour effectuer des tests établis selon des critères programmés en vue de délivrer des informations permettant le pilotage de la caméra pour la visualisation de la portion de contour contiguë.This series of information representative of the plot of the cut from the analysis processor 8 is delivered through transmission means 9 to a pilot processor 10. This processor 10 is adapted to memorize the position of said points of the plot and to perform tests established according to programmed criteria with a view to delivering information allowing the control of the camera for viewing the portion of contiguous contour.

Après réalisation des tests, le processeur pilote 10 délivre les informations de pilotage de la caméra 4 vers les moyens d'entraînement 11 des supports 2 - 3 de cette caméra et délivre également un signal de synchronisation vers le convertisseur 7 autorisant la saisie d'une nouvelle image.After carrying out the tests, the pilot processor 10 delivers the control information from the camera 4 to the drive means 11 of the supports 2 - 3 of this camera and also delivers a synchronization signal to the converter 7 authorizing the entry of a new image.

L'unité de conversion analogique/numérique 7 reçoit le signal vidéo issu de la caméra 5 et le signal de synchronisation issu du processeur pilote 10. Cette unité 7 saisit donc le signal vidéo de l'image contiguë à l'image précédemment traitée et affecte aux points de cette image des codes à plusieurs bits représentatifs des niveaux de gris de ces points parmi un nombre de niveaux prédéterminés entre le blanc et le noir, délivrant ainsi une matrice de codes numériques représentative de ladite image.The analog / digital conversion unit 7 receives the video signal from the camera 5 and the synchronization signal from the pilot processor 10. This unit 7 therefore captures the video signal from the image contiguous to the previously processed image and affects at the points of this image of the multi-bit codes representative of the gray levels of these points among a number of predetermined levels between white and black, thus delivering a matrix of digital codes representative of said image.

Cette matrice de codes numériques est délivrée vers le processeur d'analyse 8 comprenant une mémoire de stockage 12 apte à mémoriser ladite matrice de codes numériques sur commande d'une unité d'accès direct mémoire 13, une unité de calcul 14 composée d'un microprocesseur, des mémoires programmes 15 et une mémoire de travail 16 associées au microprocesseur.This matrix of digital codes is delivered to the analysis processor 8 comprising a storage memory 12 capable of storing said matrix of digital codes on command of a direct memory access unit 13, a calculation unit 14 composed of a micropro interrupter, program memories 15 and a working memory 16 associated with the microprocessor.

Les mémoires programmes 15 contiennent le programme des opérations successives à effectuer par le microprocesseur à partir de la matrice de codes numériques contenue dans la mémoire de stockage 12.The program memories 15 contain the program of the successive operations to be carried out by the microprocessor from the matrix of digital codes contained in the storage memory 12.

La première opération consiste à comparer le niveau de gris de chaque code numérique avec le niveau de gris d'un seuil déterminé, soit expérimentalement, soit lors de l'initialisation du système en générant l'image d'un objet connu. Cette opération conduit à l'obtention d'une matrice de codés binaires à partir de laquelle il est aisé de déterminer les points de l'image appartenant au fond et les points de l'image appartenant à l'objet.The first operation consists in comparing the gray level of each numerical code with the gray level of a determined threshold, either experimentally or during the initialization of the system by generating the image of a known object. This operation leads to obtaining a binary coded matrix from which it is easy to determine the points of the image belonging to the background and the points of the image belonging to the object.

La deuxième opération consiste à réaliser le lissage du contour de l'objet. Cette opération de lissage consiste à traiter tous les points de l'image par des méthodes morphologiques telles que: érosion et dilatation, permettant d'harmoniser le contour en supprimant les irrégularités de celui-ci.The second operation consists in smoothing the outline of the object. This smoothing operation consists in treating all the points of the image by morphological methods such as: erosion and dilation, making it possible to harmonize the contour by removing the irregularities from it.

Il est à noter que ces méthodes d'érosion et de dilatation sont aisément adaptables aux exigences de découpe de chaque utilisateur sans modification de la structure du dispositif. En effet, les caractéristiques du tracé de la découpe dépendent essentiellement du nombre d'érosions et de dilatations auxquelles sont soumis les points appartenant au contour de l'objet.It should be noted that these erosion and expansion methods are easily adaptable to the cutting requirements of each user without modifying the structure of the device. Indeed, the characteristics of the outline of the cut essentially depend on the number of erosions and expansions to which the points belonging to the outline of the object are subjected.

L'opération suivante consiste à extraire la position des points représentatifs du tracé de la découpe et à procéder à l'ordonnancement desdits points en vue de les restituer dans l'ordre de parcours du contour.The following operation consists in extracting the position of the points representative of the plot of the cut and in ordering the ordering of said points in order to restore them in the order of travel of the contour.

L'extraction des points du contour consiste à déterminer les points dont au moins un point voisin sur la même ligne ou la même colonne appartient au fond. Leur ordonnancement consiste à réaliser le suivi du tracé de la découpe en partant des bords de l'image et à suivre ledit tracé jusqu'à une autre limite de l'image. Cette opération est renouvelée tant qu'il reste des points du tracé non traités, en bordure d'image.The extraction of the points of the contour consists in determining the points of which at least one neighboring point on the same line or the same column belongs to the background. Their scheduling consists in carrying out the follow-up of the plot of the cutout starting from the edges of the image and in following said plot up to another limit of the image. This operation is repeated as long as there are untreated points of the plot, at the edge of the image.

Les contours non repérés représentent des trous situés à l'intérieur de l'objet, à une distance du bord de cette peau suffisamment importante pour que les opérations d'érosion et de dilatation successives ne l'aient pas éliminé.The unmarked contours represent holes located inside the object, at a distance from the edge of this skin large enough that the successive erosion and expansion operations have not eliminated it.

Il est à noter que le traitement de chaque matrice de codes numériques est réalisé en temps réel pendant le laps de temps séparant la fin de leur mémorisation dans la mémoire de stockage 12 et la saisie d'une nouvelle image dans l'unité de conversion 7. Par exemple, le temps de traitement d'une image réalisant la binarisation, puis trois érosions et dilatations, puis l'extraction de contour et enfin le suivi de ce contour, est de l'ordre de trois secondes.It should be noted that the processing of each matrix of digital codes is carried out in real time during the period of time separating the end of their storage in the storage memory 12 and the capture of a new image in the conversion unit 7 For example, the processing time of an image performing binarization, then three erosions and dilations, then the contour extraction and finally the tracking of this contour, is of the order of three seconds.

En outre, les différentes opérations de binarisation, érosion, dilatation, extraction de contours, peuvent être réalisées au moyen de modules cablés, chacun utilisé spécifiquement pour une opération déterminée, qui permettent d'obtenir des temps de calcul encore réduits.In addition, the various binarization, erosion, expansion, contour extraction operations can be carried out by means of wired modules, each used specifically for a specific operation, which allow the calculation times to be further reduced.

Ces modules câblés sont disposés en série et réalisent le traitement des codes numériques délivrés par l'unité de conversion 7 analogique /numérique, de façon à délivrer au processeur pilote 10 des informations représentatives du tracé de la découpe.These wired modules are arranged in series and carry out the processing of the digital codes delivered by the analog / digital conversion unit 7, so as to deliver to the pilot processor 10 information representative of the plot of the cut.

Un dispositif doté de modules câblés ne possède donc pas de processeur d'analyse, et les informations sur le tracé de la découpe sont transmises directement au processeur pilote qui possède dans ce cas une mémoire de stockage apte à mémoriser lesdites informations sur commande d'une unité d'accès direct mémoire.A device fitted with wired modules therefore does not have an analysis processor, and the information on the cutting plot is transmitted directly to the pilot processor which in this case has a storage memory capable of storing said information on command of a direct memory access unit.

Les informations représentatives du tracé de la découpe résultant du suivi de contour après traitement de l'image, sont ensuite délivrées vers le processeur pilote 10 dont le rôle est d'assurer la continuité dans la détermination du chemin de découpe, d'une image à l'autre, en complétant le chemin de découpe de l'image (n) avec une portion de contour de l'image (n + 1The information representative of the cutting plot resulting from the contour tracking after image processing is then delivered to the pilot processor 10 whose role is to ensure continuity in determining the cutting path, from an image to the other, by completing the image cutting path (n) with a portion of the image outline (n + 1

Ce processeur pilote 10 reçoit une série d'informations qui sont distribuées par l'entremise d'un bus de données 17 vers une unité de calcul 18 constituée d'un microprocesseur; celui-ci est associé à une mémoire programme 19 contenant les tests à effectuer et une mémoire de travail 20.This pilot processor 10 receives a series of information which is distributed via a data bus 17 to a computing unit 18 consisting of a microprocessor; this is associated with a program memory 19 containing the tests to be carried out and a working memory 20.

Afin d'assurer la continuité dans la détermination du chemin de découpe, une méthode pourrait consister à positionner le dernier point de l'image (n) au centre de l'image suivante (n+1). Toutefois cette méthode présente l'inconvénient majeur d'entraîner une deuxième analyse d'une moitié de l'image (n), ce qui ralentit notablement le processus.In order to ensure continuity in determining the cutting path, one method could consist in positioning the last point of the image (n) at the center of the next image (n + 1). However, this method has the major drawback of involving a second analysis of half of the image (n), which slows down the process considerably.

Pour remédier à cet inconvénient, la présente invention se propose de définir cinq points de repère dont la position est déterminée (figure 3a). Quatre de ces cinq points PN, PE, PS et PW sont situés sur la périphérie de l'image, chacun au centre d'un des côtés de ladite image. Le cinquième point PC est positionné au centre de l'image.To overcome this drawback, the present invention proposes to define five benchmarks whose position is determined (Figure 3a). Four of these five points PN, PE, PS and PW are located on the periphery of the image, each in the center of one of the sides of said image. The fifth point PC is positioned in the center of the image.

Le rôle du processeur pilote 10 consiste à commander un déplacement de la caméra 5 de façon que le point final Fn de l'image (n) vienne se positionner sur l'un des points de repère PN, PE, PS, PW ou PS dans l'image suivante (n + 1), de sorte que la portion de contour déjà traitée dans l'image (n) et appartenant aussi à l'image (n+1) soit minimale et la portion de contour traitée dans l'image (n + 1) maximale.The role of the pilot processor 10 consists in controlling a movement of the camera 5 so that the end point Fn of the image (n) is positioned on one of the reference points PN, PE, PS, PW or PS in the following image (n + 1), so that the portion of the contour already processed in the image (n) and also belonging to the image (n + 1) is minimal and the portion of the contour processed in the image (n + 1) maximum.

Ainsi, excepté pour le traitement de la première image, le point de début D du contour de chaque image se trouve positionné sur un point de repère PN, PE, PS, PW, PC.Thus, except for the processing of the first image, the start point D of the contour of each image is positioned on a reference point PN, PE, PS, PW, PC.

En outre, il est à noter que lorsque dans l'image (n + 1) le processeur d'analyse 8 extrait plusieurs contours distincts, le processeur pilote 10 détermine lequel desdits contours se raccorde à l'image (n) en cherchant celui qui possède un point positionné sur ou à proximité du point de repère sur lequel est situé le point final Fn de l'image (n) dans l'image (n + 1).In addition, it should be noted that when in the image (n + 1) the analysis processor 8 extracts several distinct contours, the pilot processor 10 determines which of said contours is connected to the image (n) by searching for the one which has a point positioned on or near the point of coordinate system on which the end point Fn of the image (n) is located in the image (n + 1).

Chaque image est également partitionnée en neuf régions R1...R9 similaires (figure 3b) qui permettent de prendre en compte la position du point final Fn du contour dans l'image (n).Each image is also partitioned into nine similar regions R1 ... R9 (FIG. 3b) which make it possible to take into account the position of the end point Fn of the contour in the image (n).

Le pilotage de la caméra en vue de la visualisation d'une image (n+1) tient donc compte de la position du point initial on et respectivement du point final Fn du chemin de découpe dans l'image (n) et consiste à déterminer, à partir desdites positions et en fonction de critères mémorisés, le point de repère sur lequel sera positionné le point final Fn dans l'image (n +1The control of the camera with a view to viewing an image (n + 1) therefore takes account of the position of the initial point on and respectively of the end point Fn of the cutting path in the image (n) and consists in determining , from said positions and according to memorized criteria, the reference point on which the end point Fn will be positioned in the image (n +1

Cette méthode permet avec des critères de choix judicieusement définis, de diminuer de 30 % environ le nombre d'images nécessaire à la saisie de la totalité du contour d'un objet par rapport à une stratégie plaçant systématiquement le point final Fn au centre de l'image (n + 1).This method makes it possible, with judiciously defined selection criteria, to reduce by around 30% the number of images necessary for capturing the entire outline of an object compared to a strategy systematically placing the end point Fn at the center of the 'image (n + 1).

Toutefois, cette méthode ne prenant en compte que des paramètres relatifs à la portion de peau visualisée, peut conduire à analyser des images, dites "non valides", qui comportent une très faible longueur de contour inférieure à une valeur seuil déterminée. Ceci est le cas notamment lors de l'opération de "contournage" d'une peau lorsque la caméra est positionnée au-dessus d'appendices telles que les pattes de cette peau.However, this method taking into account only parameters relating to the portion of skin viewed, can lead to the analysis of images, called "invalid" images, which have a very short contour length less than a determined threshold value. This is the case in particular during the operation of "contouring" of a skin when the camera is positioned above appendages such as the legs of this skin.

Les critères utilisés ci-dessus peuvent alors conduire à un déplacement de la caméra ne permettant pas de visualiser une portion de contour intéressante. A cet éffet, un deuxième critère est introduit qui tient compte, d'une part, du point de départ de l'analyse de la peau (ou de tout objet-plan de forme définie et présentant des appendices) et, d'autre part, du sens de parcours (trigonométrique ou inverse) (figure 3c).The criteria used above can then lead to a displacement of the camera which does not make it possible to view an interesting portion of the contour. To this effect, a second criterion is introduced which takes into account, on the one hand, the starting point of the analysis of the skin (or of any plane object of defined shape and having appendages) and, on the other hand , direction of travel (trigonometric or reverse) (Figure 3c).

Le processeur pilote mémorise les coordonnées en X et Y de la caméra 5 sur la table de déplacement 1 et en fonction des coordonnées du point de départ et du sens de parcours, peut ainsi déterminer, lors de la visualisation d'une image "non valide", si cette image correspond ou non à une région particulière de la peau.The pilot processor stores the coordinates in X and Y of the camera 5 on the movement table 1 and as a function of the coordinates of the starting point and the direction of travel, can thus determine, when viewing an "invalid" image ", whether or not this image corresponds to a particular region of the skin.

Dans l'affirmative, le processeur inhibe les résultats des tests prenant en compte les positions du point de début et du point final du contour de l'image (n) et détermine la position du point final de l'image (n) dans l'image (n + 1) en fonction de la région où est positionnée la caméra.If so, the processor inhibits the results of the tests taking into account the positions of the start point and the end point of the outline of the image (n) and determines the position of the end point of the image (n) in l image (n + 1) depending on the region where the camera is positioned.

Il est à noter que lorsque ces critères conduisent eux-mêmes à la visualisation d'une deuxième image "non valide", les résultats des tests sont inhibés et le point final du contour de l'image (n) est systématiquement positionné au centre PC de l'image (n+1). Cette solution est également adoptée lors de la visualisation d'une image non valide en dehors des régions particulières prédéterminées.It should be noted that when these criteria themselves lead to the display of a second "invalid" image, the test results are inhibited and the end point of the image outline (n) is systematically positioned in the PC center of the image (n + 1). This solution is also adopted when viewing an invalid image outside the particular predetermined regions.

La figure 4 représente à cet effet toutes les combinaisons permettant de déterminer la position du point final Fn de l'image (n), dans l'image (n+1).FIG. 4 represents for this purpose all the combinations making it possible to determine the position of the end point Fn of the image (n), in the image (n + 1).

Lorsque la totalité du contour de l'objet est visualisée, les informations mémorisées dans la mémoire de travail 20 du processeur pilote 10 sont délivrées par l'intermédiaire d'interfaces de communication 21 à des moyens de découpe 22 industriels tels que ciseaux électriques, laser, jet de fluide, aptes à réaliser la découpe effective de la peau.When the entire contour of the object is viewed, the information stored in the working memory 20 of the pilot processor 10 is delivered via communication interfaces 21 to industrial cutting means 22 such as electric scissors, laser , jet of fluid, capable of carrying out the actual cutting of the skin.

Les figures 5a et 5b représentent une peau avant et après traitement et l'on peut noter que cette peau présente après traitement un contour beaucoup plus régulier.FIGS. 5a and 5b represent a skin before and after treatment and it can be noted that this skin has, after treatment, a much more regular outline.

Un dispositif, tel que décrit ci-dessus, et mettant en oeuvre des modules câblés afin de réaliser les opérations de binarisation, érosion, dilatation, extraction de contours, permet de réaliser la saisie du contour d'une peau de mouton, et à déterminer le chemin optimal de découpe avec une cadence de deux à trois peaux à la minute. Ce rendement peut en outre être amélioré en disposant deux caméras mobiles, chacune analysant une partie de la peau.A device, as described above, and using wired modules in order to carry out the binarization, erosion, expansion, contour extraction operations, makes it possible to capture the outline of a sheep skin, and to determine the optimal cutting path with a rate of two to three skins per minute. This performance can also be improved by having two mobile cameras, each analyzing part of the skin.

Claims (8)

1. Method for determining the trace for cutting the contour of a plane object as a function of parameters relating to the peripheral aspect of said object, of the type consisting:
- in generating, by means of a video camera (5), a screened image of a part of the contour of the object (6) consisting of points distributed in lines and columns,
- in capturing, in real time, the analog signal of said image, in converting it to a matrix of digital codes, each representative of the degree of greyness of a point of the image, and in storing said matrix in real time, said process being characterised in that it consists:
- in processing, by calculation, the matrix of digital codes as a function of the cutting parameters, so as to determine the points representative of the cutting trace,
- in storing the data representative of the position of the points representative of the cutting trace,
- in subjecting said data to a series of tests laid down in accordance with programmed criteria, with a view to supplying data enabling the camera to be guided in order to visualise the adjacent section of the contour,
- and in controlling actual cutting, once the entire contour has been stored.
2. Method according to claim 1, characterised in that, for each image (n) processed, it consists:
- in proceeding to sequence the stored data, representative of the position of the points of the cutting trace, with a view to restring them to the order of the path of the contour,
- in extracting from said data the data representative of the position of the point of origin (Dn) of the contour section, and of the end point (FN) of said contour section,
whereby the tests to which said data are subjected consist in comparing the position of the point of origin (Dn) and of the end point (FN) of said image (n), respectively, with the stored reference values, the position of which is determined, with a view to supplying information representative of the position of the end point (Fn) of image (n), in image (n+1).
3. Method according to claim 2, characterised in that it consists, on the one hand, in defining a given number of reference points, the position of which is predetermined, and, on the other hand, in dividing each image into a number of specific zones, whereby the tests to which the data representative of the position of the point of origin and of the end point of the contour section of an image (n) are subjected, consist, on the one hand, in determining the reference point, the distance of which from the point of origin Dn is minimal, and, on the other hand, in determining the zone of image (n), in which the end point (Fn) is located, and in deducing therefrom, in accordance with programmed criteria, the position of said end point (Fn) in the adjacent image (n + 1).
4. Method according to claim 3, characterised in that it consists in locating the end point (Fn) of image (n) on a reference point, so that it constitutes the point of origin of the contour section of image (n+1).
5. Method according to one of the preceding claims, characterised in that processing of the matrix of digital codes consists:
- in digitalising said matrix of digital codes in the form of binary codes, the value of which is a function of the presence or absence of the object, with a view to obtaining data representative of the contour of the object,
- in processing said data representative of the contour of the object by morphological methods such as erosion, expansion... with a view to supplying data representative of the position of points of the cutting trace.
6. Device for cutting the contour of a plane object with a view to implementing the method in accordance with one of the preceding claims, of the type comprising:
- at least one video camera (5) capable of generating successively video images of ccntour sections of the object (6),
- an analog/digital converter (7) receiving the video signal from camera (5) and capable of ensuring capture of said signal and supply of a matrix of digital codes depending on the level of luminance of the sampled video signal,
- a module (8) referred to as the "analysis processor" and comprising:
. a storage memory (12) capable of storing each matrix of digital codes and controlled by a direct-access memory (13) capable of managing the arrangement of digital codes in the memory,
. a calculating unit (14) capable of ensuring that the digital codes contained in storage memory (12) are processed and linked, on the one hand, with a program memory (is) containing a program of logical operations to be carried out, and, on the other hand, to a working memory (16) for provisional storage of the results,
- a module (10) referred to as the "guide processor" and designed to receive the data emanating from analysis processor (8) and comprising a calculating unit (18) capable of subjecting the data received to tests and linked, on the one hand, to a program memory (19) containing the tests to be carried out, and, on the other hand, to a working memory (20) for storing the data emanating from the analysis processor,
- transmission means (9) capable of transmitting the data from analysis processor (8) to guide processor (10),
said device being characterised in that it comprises:
- means for displacing (11) each camera (5) capable of bringing about level displacement of said cameras,
- a communication interface (21) for data in accordance with the result of tests, emanating from the guide processor (10) towards displacing means (11) of each camera, with a view to guiding the displacement of the latter,
- a communication interface (21) for the data relating to the cutting trace, emanating from guide processor (10) towards the means for cutting (22) of the contour of the plane object.
7. Device according to claim 6, of the type comprising a "C.C.D." charge transfer circuit, with a view to generating video images of successive sections of the contour.
EP19860905837 1985-09-27 1986-09-23 Method and device for cutting the contour of a plane object Expired EP0237555B1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
FR8514970A FR2588102B1 (en) 1985-09-27 1985-09-27 METHOD AND DEVICE FOR CUTTING THE CONTOUR OF A PLANAR OBJECT
FR8514970 1985-09-27

Publications (2)

Publication Number Publication Date
EP0237555A1 EP0237555A1 (en) 1987-09-23
EP0237555B1 true EP0237555B1 (en) 1989-07-12

Family

ID=9323678

Family Applications (1)

Application Number Title Priority Date Filing Date
EP19860905837 Expired EP0237555B1 (en) 1985-09-27 1986-09-23 Method and device for cutting the contour of a plane object

Country Status (6)

Country Link
EP (1) EP0237555B1 (en)
ES (1) ES2003135A6 (en)
FR (1) FR2588102B1 (en)
GR (1) GR862453B (en)
PT (1) PT83448B (en)
WO (1) WO1987002151A1 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109762949A (en) * 2019-03-18 2019-05-17 香港纺织及成衣研发中心有限公司 A kind of fur manufacturing method, device and equipment

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4998005A (en) * 1989-05-15 1991-03-05 General Electric Company Machine vision system
US5205232A (en) * 1989-08-30 1993-04-27 Orisol Ltd. Apparatus for advance edge detection and sewing
FI20021138A0 (en) * 2002-06-12 2002-06-12 Kvaerner Masa Yards Oy Procedure and arrangement for processing one or more objects
CN113911629A (en) * 2021-10-19 2022-01-11 精基科技有限公司 Intelligent coal flow conveying system
CN116984956B (en) * 2023-09-05 2024-05-03 广州亿达科技有限公司 Intelligent machining method and system for cutter grinding machining center

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE3138364A1 (en) * 1981-09-26 1983-04-07 Männel, Friedrich, 7527 Kraichtal "METHOD FOR CONTROLLING AN EMBROIDERY MACHINE"

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109762949A (en) * 2019-03-18 2019-05-17 香港纺织及成衣研发中心有限公司 A kind of fur manufacturing method, device and equipment
CN109762949B (en) * 2019-03-18 2021-04-27 香港纺织及成衣研发中心有限公司 Fur processing method, device and equipment

Also Published As

Publication number Publication date
EP0237555A1 (en) 1987-09-23
PT83448B (en) 1992-10-30
FR2588102B1 (en) 1987-11-27
FR2588102A1 (en) 1987-04-03
ES2003135A6 (en) 1988-10-16
GR862453B (en) 1987-01-27
PT83448A (en) 1986-10-01
WO1987002151A1 (en) 1987-04-09

Similar Documents

Publication Publication Date Title
EP0243253B1 (en) Method for automatically recognizing objects susceptible to overlapping one another
FR2549259A1 (en) CHARACTER RECOGNIZING METHOD AND DEVICE
FR2665597A1 (en) METHOD AND DEVICE FOR REAL TIME LOCALIZATION OF RECTILINEED CONTOURS IN A DIGITIZED IMAGE, IN PARTICULAR FOR THE RECOGNITION OF FORMS IN A SCENE ANALYSIS PROCESS.
FR2611287A1 (en) METHOD AND DEVICE FOR IDENTIFYING PARTS DURING MANUFACTURE
FR2751772A1 (en) METHOD AND DEVICE OPERATING IN REAL TIME FOR LOCALIZATION AND LOCATION OF A RELATIVE MOTION AREA IN A SCENE, AS WELL AS FOR DETERMINING THE SPEED AND DIRECTION OF MOVEMENT
EP0237555B1 (en) Method and device for cutting the contour of a plane object
FR2533040A1 (en) METHOD FOR COMPRESSING DATA
FR2645782A1 (en) SYSTEM FOR CONTROLLING TOOL BREAK ON MACHINING CENTER
EP0435768B1 (en) Method and device of attitude measurement of a moving body by tracking a pattern placed on the body
EP4046129B1 (en) Manufacturing method
EP0211789A1 (en) Process for automatically cutting a hide or the like by placing models on this hide, and apparatus for carrying it into effect
FR2772225A1 (en) PROCESS FOR PROCESSING AN IMAGE WHERE NOISE DEPENDS ON THE SIGNAL
EP0276603B1 (en) Digital signal coding and transmission method by way of timed selective refreshing of a vectorial quantizer
CN116958137B (en) New energy battery surface defect detection method and device
JP2006048205A (en) Pupil detecting device and pupil authenticating device
EP0926625B1 (en) Image processing method incuding spatio-temporal noise filtering steps and medical imaging apparatus applying this method
FR2611063A1 (en) Method and device for real-time processing of a sequenced data flow, and application to the processing of digital video signals representing a video image
CN117392178A (en) Method and device for extracting motion characteristics of molten pool in powder spreading and material adding manufacturing process
CA2296662A1 (en) Optoelectronic acquisition device for one- and two-dimensional codes
EP2022009B1 (en) Method of coding and system for displaying on a screen a numerical mock-up of an object in the form of a synthesis image
EP0076209B1 (en) Automatic device for the statistical analysis of an object
FR2583545A1 (en) IMPROVING METHODS AND DEVICES FOR AUTOMATICALLY DIGITIZING A SCENE COMPRISING DISCRETE SIGNIFICANT ELEMENTS
FR2622968A1 (en) METHOD AND DEVICE FOR PERFORMING DIMENSIONAL CONTROL OF A PART BY OPTOELECTRONIC ROUTE
EP0385528B1 (en) Process and system for the determination of the position and relative orientation of two objects in space
EP0363542A1 (en) Videogrammetic device and process

Legal Events

Date Code Title Description
PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

AK Designated contracting states

Kind code of ref document: A1

Designated state(s): IT

17P Request for examination filed

Effective date: 19870817

17Q First examination report despatched

Effective date: 19881214

GRAA (expected) grant

Free format text: ORIGINAL CODE: 0009210

AK Designated contracting states

Kind code of ref document: B1

Designated state(s): IT

ITF It: translation for a ep patent filed

Owner name: ORGANIZZAZIONE D'AGOSTINI

PLBI Opposition filed

Free format text: ORIGINAL CODE: 0009260

26 Opposition filed

Opponent name: MESSER GRIESHEIM GMBH

Effective date: 19900411

RDAG Patent revoked

Free format text: ORIGINAL CODE: 0009271

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: PATENT REVOKED

27W Patent revoked

Effective date: 19910930